English as North Germanic (original) (raw)

Abstract

The present article is a summary of the book English: The Language of the Vikings by Joseph E. Emonds and Jan Terje Faarlund. The major claim of the book and of this article is that there are lexical and, above all, syntactic arguments in favor of considering Middle and Modern English as descending from the North Germanic language spoken by the Scandinavian population in the East and North of England prior to the Norman Conquest, rather than from the West Germanic Old English.

1. Introduction

The forerunner of Modern English is the 14th-century Middle English dialect spoken in Britain’s East Midlands (Baugh and Cable, 2002: 192–193; Pyles, 1971: 155–158). All available evidence thus indicates that the ancestor of today’s Standard English is the Middle English of what before the Norman Conquest (1066) was called the Danelaw. The texts in this dialect have a recognizable syntax that separates them from a different and also identifiable Middle English system, broadly termed ‘southern.’ In our book English: The Language of the Vikings (Emonds and Faarlund, 2014), we try to determine the synchronic nature and historic source of this East Midlands version of Middle English, which then also reveals the source of Modern English. The lexicon alone cannot be used to decide these questions. What must instead determine the conclusions is the grammatical system, specifically the syntactic constructions.

As elements from other domains of language, syntax may of course also be borrowed under certain circumstances, but it happens to a much lesser extent than is the case with vocabulary. Syntactic continuity and stability through history is therefore a clear indication of a genealogical linguistic relationship. In our context it is significant that the syntax of Middle and Modern English is much more Scandinavian in nature than it is French, despite the strong and long-lasting Norman influence on the English lexicon.

In the book, we show that both synchronically and historically, Middle (and Modern) English is unmistakably North Germanic and not West Germanic. (Uncontroversially, Old English, just like Dutch and German, is West Germanic.) That is, Middle English did not develop from Old English. Old English is the language of mainly West Saxon texts, of which the last exemplars are widely taken to be the earlier Peterborough Chronicles through 1121 (Freeborn, 1998: 82). We claim that Middle and Modern English are instead direct descendants of the language spoken by Scandinavians who had relocated to England over more than two centuries prior to the Norman Conquest. We refer to this earlier language as Norse.1 Since this language, over time, acquired a large component of words from Old English, we may use the synonym “Anglicized Norse” for the early Middle English of the East Midlands spoken and written in the 12th and 13th centuries.

William the Conqueror and his French-speaking Norman armies overran and completely subdued all of England in 1066 and the decades following. As a result of this, both the English and Scandinavians were thoroughly dispossessed. The miserable circumstances gave rise to a complete fusion of two previously separate populations, speakers of Old English and speakers of Norse. The two distinct linguistic communities of the time before 1066 ended up speaking a single language by, say, 1300, i.e., what is today called Middle English. The change from two languages to one in the 12th and 13th centuries is not simply the merging of two highly similar systems. All sources agree that Middle English has great numbers of both Old English and Norse words (in addition, Old English and Norse had a high percentage of mutually comprehensible cognates). In light of the syntactic arguments we present, there are only two plausible ways to describe what happened during the two centuries after the Norman Conquest:

1. I. The traditional scenario: Middle English developed from Old English. Old English underwent many fundamental grammatical changes, incorporated much Norse vocabulary, and became Middle English.
2. II. Our alternative scenario: Middle English developed from Norse. Norse underwent essentially no grammatical changes other than those initiated on the Mainland, incorporated somewhat more Old English vocabulary, and became Middle English.

2. The Lexicon

Claims about genealogical linguistic relationships can be based on only the most primitive parts of the lexicon (small numbers, kinship terms, basic physical items). These items in Middle English, Norse, and Old English are almost all obvious cognates, and therefore irrelevant for deciding between the above scenarios.

Nonetheless, there are other aspects of the Old and Middle English lexicons that, at least to some extent, may support our claim. First, according to calculations based on word lists in Freeborn (1992) and Baugh and Cable (2002), about half of all the Germanic words of Middle English are common Germanic cognates. Traditionally, they have been counted as continuations of Old English, but they may just as well be Norse.2

Second, the character of the Norse words in Middle English is telling. Although the majority of the non-cognate Germanic words may be from Old English (perhaps 2/3 of them), the Norse words are typically daily-life words, words for objects and concepts that Old English also must have had. We mention just a few typical examples out of hundreds: bag, birth, both, call, crook, die, dirt, dike, egg, fellow, get, give, guess, likely, link, low, nag, odd, root, rotten, sack, same, scrape, sister, skin, skirt, sky, take, though, ugly, want, wing, etc. It is essentially unheard of that a living language on its own territory borrows huge numbers of daily-life terms from an immigrant population whose language dies out, yet that is what the traditional scenario is forced to claim about Middle English. Burnley (1992), in fact, concludes that about half the common Germanic words of English are not of English origin, and very few of these, relatively speaking, have any source other than Scandinavian.

Third, our research shows that the Middle English grammatical lexicon is composed in nearly equal percentages of Scandinavian and Old English elements. For example, the personal pronouns are evenly distributed between the two sources: First and second person pronouns are common Germanic, third person singular he/him is from Old English, she is of uncertain origin, and third person plural they/them is from Norse. Another example is provided by the English grammatical verbs (are, be, come, do, get, go, have, let, were, etc.), which includes slightly more items from Norse than from Old English. Such fusion of grammatical vocabulary is highly unusual in the histories of languages, and in itself supports neither the traditional scenario nor ours. We must look elsewhere for evidence to decide between the scenarios—in the syntax. The main part of our argument is based on the syntactic structures of West and North Germanic, specifically Middle English, Norse, and Old English.

In typical cases, even when massive lexical borrowing is underway, native speakers maintain their syntax. Second-language learners eventually adapt and, after a few generations at most, adopt a more or less unchanged native syntax. Under our scenario II above, this is exactly what happened in the Danelaw/East Midlands. In our book, we show that the grammar maintained was decisively Norse. The second-language learners living among native speakers were, therefore, those speaking Old English.

In the following section, we treat a series of syntactic patterns and phenomena that Middle English (and in most cases Modern English) shares with Norse, but not with Old English or present-day West Germanic languages. Our overall argument that Middle English descends from Norse depends not on its exhibiting any one specific syntactic feature of Norse, but rather on Middle English having so many of these features, while at the same time exhibiting essentially no Old English characteristics not shared by Norse.

3. Syntactic Properties of Norse and Middle English Not Shared by Old English

3.1. Word Order in Verb Phrases

All the Germanic languages at their medieval stages exhibit a considerable variability in their verb-complement order, unlike most contemporary standard varieties. Modern German and Dutch are verb-final (OV), while Modern English sides with Norse in its consistent VO pattern.

Although Old English data are to some extent inconclusive and have triggered conflicting analyses (Fuss and Trips, 2002; Pintzuk, 2002a, b), earlier extensive research of both generative (van Kemenade, 1987; Roberts, 1997) and non-generative inspiration (Stockwell and Minkova, 1991) concluded that the basic word order of Old English verb phrases largely conformed to that of the modern West Germanic languages; that is, the dominant underlying Old English word order in the VP was V-final, with the verb-second position in main clauses being due to movement. Norse had changed to an underlying VO order by the 9th century (Faarlund, 2002: 949; 2004: 160–166), although some variation, with some marked cases of OV order, is found later than that.

At the start of the Middle English period, between 1150 and 1250, there seemed to be an abrupt change. Pintzuk and Taylor (2006) provide figures showing a sharply declining frequency of head-final VP structures in the Middle English period, indicating that English was robustly a VO language after 1250: 97 % of examples show this word order.

3.2. From Verbal Prefixes to Post-verbal Particles

Like other West Germanic languages, Old English had a productive system of directional and aspectual verbal prefixes, such as the perfective prefixes ge- and be-. The extensive standard Old English pattern of these particles is like the rest of West Germanic.

This system of prefixes became much less prominent in the Middle English period, and in fact died out as a productive pattern. On the other hand, Middle English rather suddenly developed a system of post-verbal particles which took over the role of the prefixes.

Directional and aspectual particles were generally post-verbal, not only in main clauses but in subordinate clauses with both finite and non-finite verbs, as in today’s English.

North Germanic had lost the verbal prefixes in prehistoric times, and replaced them with a post-verbal system that is exactly what appeared in Middle English, as in 13th-century Danish:

Lamont (2005) summarizes ten authors who treat the plentiful numbers of “phrasal verbs” introduced into early Middle English, and who essentially concur that Old English lacked such collocations. One (Hiltunen, 1983: 92) concludes, “… one cannot avoid the impression of the prefixes having been swept away almost overnight. The suddenness of the change is remarkable in view of the longish and stable OE period.” From our perspective, the Middle English system of post-verbal particles simply continues the Norse pattern.

3.3. Subject-to-Subject Raising

A familiar pattern in Modern English is so-called subject-to-subject raising, whereby the subject of a subordinate clause may occur on the surface as the subject of the matrix clause, as in You seem to know the answer. Subject raising of this kind is absent from Old English: “Unquestionable instances of subject-raising with verbs like _þync_- ‘seem’ are hard to find” (Traugott, 1972: 102). “Subject Raising was rare before the second half of the Middle English period” (Denison, 1993: 221). “Old English had very similar Raising and Tough Movement possibilities to those of Modern German, i.e. effectively no S[ubj]-S[ubj] or S[ubj]-O[bj] Raising” (Hawkins, 1986: 82).

In contrast, with the Norse verb þykkja ‘seem’ and other verbs with a similar meaning, subject raising is the normal and unmarked construction.

In (3) hann is the grammatical subject of þótti ‘seemed,’ but the underlying subject of the infinitive phrase vera …

Middle English definitely exhibits subject raising, well before Chaucer.

In (4), war & wirrsenn is the grammatical subject of toc (past tense of taka, a Norse word), while it is the underlying subject of flowenn in the subordinate clause. In (5), lickli behaves just like the modern likely, raising the subject of the subordinate clause, here yee ‘you,’ to become its surface subject.

3.4. Subject-to-Object Raising

(Also known as Accusative with infinitive, Exceptional Case Marking (ECM) or Raising to Object.) In this configuration, what would be the subject of a subordinate clause appears to be the object of a matrix verb, in that it receives (accusative) case from it and can be a reflexive pronoun bound by a higher subject: Do you want her to speak louder? (This raising is not to be confused with so-called “small clauses” that appear with perception verbs: I heard John cough.)

True Subject-to-Object Raising (ECM) is absolutely absent from Old English: “in Latin translation, Old English and Old German had S-O raising structures, but these were calques from Latin” (Auwera and Noël, 2011: 22; and see again Hawkins, 1986). It was, however, very common in Norse, see (6), where the subordinate clause consists of the accusative subject guð yðarn and the infinitive predicate gera. Eventually, Middle English exhibited equivalent constructions, although in (7) the accusative case is no longer visible.

We conclude that the introduction of raising constructions is not a syntactic “change” that must be postulated in the history of Middle English. Rather, raising, though a construction somewhat literate in tone, is simply a device continuously available in Norse syntax, used to different degrees by different authors and in different styles of Middle English.

3.5. Preposition Stranding

Prepositions can be stranded both through A-movement (passive) and A’-movement (_wh_-movement and topicalization). The possibility of both types is a unique feature of Scandinavian (Norwegian examples) and English.

8 Den ulykka vart aldri snakka om

That accident was never talked about

9 Kva snakkar du om ?

What are you talking about ?

The full array of P-stranding of this kind does not exist in any other Germanic language, including Old English. Similar-looking phenomena in Old English are of a different kind, which is not found in Middle or Modern English (van Kemenade, 1987: 153).

Topicalization with P-stranding is also found in Norse texts at least from the 13th century.

This process did not exist in Old English in any general sense. Then it suddenly shows up in 13th-century Middle English (Fischer, 1992: 389).3

It would be too much of a coincidence if such a rare construction showed up independently in two different but closely related languages at about the same time.

3.6. Split Infinitive

In Germanic, the infinitive in some of its uses (generally control infinitives) is preceded by an infinitive marker, as the English to. In English and Norwegian, an adverb may intervene between the infinitive marker (Norw. å) and the verb.

14 Det er viktig å alltid komma i tide

It is important to always come on time

In Old English, the infinitive marker was invariably adjacent to the following verb. This is still the case throughout West Germanic, e.g. for Dutch te and German zu. None of these languages ever tolerate split infinitives.

In Old Norse, the infinitive marker at was a complementizer (Faarlund, 2007). Since the language at that stage (unlike today’s Mainland Scandinavian) had verb raising to T even in non-finite clauses (i.e., the verb precedes the sentence adverb), and since the subject would be null in infinitival clauses, nothing would normally intervene between the infinitive marker and the verb, as in (15), cf. the structure (15’). Occasionally, however, an adverbial could be adjoined to TP and thus intervene between at and the verb, see (16).

15’ ætluðu CP [at TP [PRO hengja VP [PRO hengja hann]]]

After verb movement to T was lost, sentential adverbials began to appear more freely between the verb and the infinitive marker in both Scandinavian and English, as in (17–18). From the 14th century on, we find split infinitives in Middle English:

18 It is good to not ete fleisch and to not drynke wyn (Purvey, Romans 16:21)

‘It is good to not eat meat and to not drink wine’

Whether to is in C or in T at this stage, it is clearly separated from the verb, as in North Germanic. The split infinitive is thus another sign that, since medieval times, English has been North rather than West Germanic.

3.7. Verb-Second

Both North and West Germanic languages typically exhibit a Verb-Second (V2) property in main clauses, including declaratives, with (only) Modern English being exceptional in this regard. Although Old English had predominantly V2 structures, it allowed some Verb-Third sequences in main clauses, in which both some topicalized phrase and a pronominal subject are to the left of the finite verb.

A thorough study of the pronoun placement in Middle English (Kroch, Taylor, and Ringe, 2000) demonstrates that V2 in the texts of the East Midlands and North conforms to the standard V2 of Mainland Scandinavian. In Northern and East Midlands dialects, e.g. our Anglicized Norse, subject pronouns regularly obey the general North Germanic pattern: namely, they follow sequences of initial XP + finite verb. This typical use of V2 continues into Middle English; there is thus an unbroken continuity between Norse and Middle English. The Verb-Third patterns continue in the Middle English of the South and West Midlands as a southern continuation of Old English, but texts with traces of Old English syntax are rare after 1250 (Kroch, Taylor, and Ringe, 2000).

4. Shared Innovations

This section will present several morphosyntactic characteristics that are common to later Middle English and Norse, excluding Old English and other West Germanic languages, but whose Norse roots are not easily found in written texts. The basic reason for this lack is that, until after the Conquest, neither Northern English dialects nor Anglicized Norse were written languages. So, when we find innovative features common to Anglicized Norse and Mainland Scandinavian, our arguments cannot, in principle, be expected to be based on robust textual evidence, any more than evidence for Proto-Romance (≠ Latin) or Proto-Slavic (≠ Old Church Slavonic) can. Rather, we must use a time-honored method of diachronic linguistics, syntactic reconstruction. That is, if hypothesized daughter languages, here Middle English and Mainland Scandinavian, share some unusual and rare innovative characteristic, then it is justified to conclude that the source is in the parent of both, i.e. Old Norse.

4.1. The Genitive Suffix

As the case system of Norse eroded, the genitive case suffix -s was reanalyzed as a phrasal clitic, as in the Danish in (19).

This reanalysis has never occurred in a West Germanic language. This Mainland Scandinavian pattern is exactly that of Middle and Modern English, as seen in the glosses. Similarly, the Middle English phrase (20) has the genitive case marker only on the last noun.

20 þurh þe Laferrd Criste ss dæþ (Orm., c. 1180)

through the Lord Christ’s death

Modern English thus shares the genitive phrase-final clitic -s of North Germanic. But Old English spells out genitive case on both head nouns and determiners, like morphological case in languages such as German.

4.2. Pronominal Object Forms

English and Mainland Scandinavian (especially Danish) extend pronominal object forms to all positions other than uncoordinated subjects of overt finite verbs (Emonds, 1986).

21 Me and John went to the movie

Mig og John gik i biografen

22 John is better qualified than them

John er bedre kvalificeret end dem

In contrast, the West Germanic (e.g., Dutch) tendency is less pronounced and goes rather in the direction of extending subject forms such as ik ‘I’ to positions where prescriptive and most adult usage requires an object form (A. van Hout, pers. comm.).

4.3. Grading of Long Adjectives

English and Modern Scandinavian are the only Germanic languages in which free words meaning ‘more’ and ‘most’ have come to be the general means for grading longer and infrequent adjectives, while the modern West Germanic languages continue the old synthetic form of gradation even of long adjectives.

23 Sie ist interessanter als ihr Mann (German)

Ho er meir interessant enn sin mann (Norwegian)

She is more interesting than her husband

4.4. Parasitic Gaps

This is the term for the phenomenon whereby a gap in one part of the sentence resulting from topicalizing or otherwise fronting a noun phrase licenses a second gap with the same reference in another part of the sentence (Engdahl, 1983). Consider (24) with an empty object position and a second empty position further to the right (the “parasitic gap”), compared to (25), which has no topicalization and so the second object is obligatorily expressed. Exactly the same pattern exists in today’s Scandinavian, as shown by the equivalent Norwegian examples (26–27).

24 That book i I returned t i without having read.

25 I returned the book without having read *(it).

Both Old English and Norse would allow object gaps in any case, parasitic or not (Mitchell and Robinson, 1992: 107; Faarlund, 2004: 166–168). Therefore we need to compare English and Scandinavian with a contemporary West Germanic language which does not allow null arguments, such as German. Speakers of Modern German do not accept sentences with parasitic gaps of this kind.

Since parasitic gaps as in (24) and (26) are characteristic of North Germanic and not West Germanic, English again patterns with the former group.

4.5. Tag Questions

English has a form of tag questions which copies the first auxiliary of the declarative, reverses the polarity of the declarative, and then inverts a pronoun copy of the subject. Moreover, if the declarative lacks an auxiliary, the tag questions must contain an agreeing form of the auxiliary do.

29 John has gone home, hasn’t he?

Sue went home, didn’t she?

English is highly unusual in its common use of this kind of tag. For example, these variable tags transliterated into the Western Germanic language German are totally ungrammatical. In fact, the only other Germanic languages to have such tag questions are North Germanic, for example Norwegian. The rules for constructing Norwegian tag questions are exactly those of English, including the required use of the ‘pro-verb’ gjøre/gjera ‘do’ with declaratives that lack an auxiliary verb.

Keeping in mind that these variable tag questions are grammatically quite complex, it is highly unlikely that they “accidentally” developed twice, only in England and in Scandinavia. They may be undetected in corpora of literary, religious and administrative documents, since they are basically a conversational device. But they remain limited to North Germanic.

5. Summary of English Syntax

The overtly signaled syntactic features and constructions of Middle and Modern English belong to one of the following categories:

Common Germanic:

Simple adjectives before noun, head-initial structures in noun phrases, impersonal existential sentences, non-finite participles, obligatory expression of plural on nouns, periphrastic tenses and aspect of verbs, prepositions (rather than postpositions), _wh_-fronting in questions, wh in headless relatives, etc.

Specifically late Middle and Modern English:

Inversion of Tense without V and _do_-support, progressive tense, use of V-ing as a participle, Verb-Third after fronted phrases and subjects, _wh_-forms as relative pronouns.

English shared with North Germanic (not West Germanic):

“Case levelling” using object pronouns, modals to express the future, perfect infinitives (with have), phrasal genitives, preposition stranding, invariant relative complementizer, split infinitives, subject-to-subject raising, subject-to-object raising, parasitic gaps, periphrastic grading of adjectives, tag questions based on syntactic copies, VOX word order, verb-particle constructions, etc.

(Only a selection of these are discussed in this short article, the rest are in the book.)

English shared with West Germanic (not North Germanic):

(We have been able to find none, even when we ask audiences to look for them.)

The syntactic evidence for the ancestor of English thus all goes one way. Therefore, by syntactic criteria, Middle and Modern English are North Germanic. In the immediate aftermath of the Norman Conquest and the social upheaval that it caused, Middle English was created by a remarkably large importation of Old English vocabulary into both the open-class and grammatical lexicons on the syntactic model of Norse. This final fusion of the two lexicons may well have been preceded by a lot of borrowing of Old English into (unwritten) Anglicized Norse before the Conquest.

6. Conclusion

Middle English speakers of the East Midlands and the North did not “borrow” Norse words; children simply brought them into their native language from the late 11th to the early 13th century by appropriating from their parents’ and peers’ Norse and Old English vocabularies on a nearly equal basis. As a grammatical system, they used the Norse model. While the parents may often have been speaking mutually comprehensible amalgams of their different native Germanic languages, their children were already creating from these vocabularies a _new North Germanic tongue_—the language which we today call Middle English.

References4

van der Auwera, Johan and Dirk Noël. 2011. Raising: Dutch between English and German. Journal of Germanic Linguistics 23: 1–36.

Baugh, Albert and Thomas Cable. 2002. A History of English. 5th ed. London: Routledge and Kegan Paul.

Burnley, David. 1992. Lexis and semantics. In Norman Blake (ed.), The Cambridge History of the English Language, Vol. 2, 409–460. Cambridge: Cambridge University Press.

Denison, David. 1993. English Historical Syntax. London: Longman.

Emonds, Joseph. 1986. Grammatically deviant prestige dialect constructions. In Michael Brame, Heles Contreras, and Frederick Newmeyer (eds.), A Festschrift for Sol Saporta, 93–129. Seattle: Noit Amrofer.

Emonds, Joseph and Jan Terje Faarlund. 2014. English: The Language of the Vikings. Olomouc: Palacký University Press. Downloadable at http://anglistika.upol.cz/vikings2014/ (accessed March 30, 2016).

Engdahl, Elisabet. 1983. Parasitic gaps. Linguistics and Philosophy 6: 5–34.

Faarlund, Jan Terje. 2002. The syntax of Old Nordic. In Oskar Bandle, Kurt Braunmuller, Ernst Hakon Jahr, Allan Karker, Hans-Peter Naumann, Ulf Telemann, Lennart Elmevik, and Gun Widmark (eds.), The Nordic Languages. An International Handbook of the History of the North Germanic Languages, Vol. 1, 940–950. Berlin: Mouton de Gruyter.

Faarlund, Jan Terje. 2004. The Syntax of Old Norse. Oxford: Oxford University Press.

Faarlund, Jan Terje. 2007. Parameterization and change in non-finite complementation. Diachronica 24: 57–80.

Fischer, Olga. 1992. Syntax. In Norman Blake (ed.), The Cambridge History of the English Language, Vol. 2, 207–408. Cambridge: Cambridge University Press.

Freeborn, Dennis. 1992. Word Book. London: MacMillan Press.

Freeborn, Dennis. 1998. From Old English to Standard English. 2nd ed. London: MacMillan Press.

Fuss, Eric and Carola Trips. 2002. Variation and change in Old and Middle English—on the validity of the Double Base Hypothesis. Journal of Comparative Germanic Linguistics 4.3: 171–224.

Hawkins, John A. 1986. A Comparative Typology of English and German: Unifying the Contrasts. London: Croom Helm.

Hiltunen, Risto. 1983. The Decline of the Prefixes and the Beginnings of the English Phrasal Verb. Turku: Turun Yliopisto.

van Kemenade, Ans. 1987. Syntactic Case and Morphological Case in the History of English. Dordrecht: Foris Publications.

Kroch, Anthony, Ann Taylor, and Donald Ringe. 2000. The Middle English verb-second constraint: A case study in language contact and language change. In Susan Herring, Pieter van Reenen, and Lene Schøsler (eds.), Textual Parameters in Older Language, 353–391. Amsterdam/Philadelphia: John Benjamins.

Lamont, George. 2005. The Historical Rise of the English Phrasal Verb. Accessible at http://homes.chass.utoronto.ca/\~cpercy/courses/6361lamont.html (accessed March 30, 2016).

Mitchell, Bruce and Fred C. Robinson. 1992. A Guide to Old English. 5th ed. Padstow: T.J. Press.

Pintzuk, Susan. 2002a. Morphological case and word order in Old English. Language Sciences 24: 381–395.

Pintzuk, Susan. 2002b. Verb-object order in Old English: Variation as grammatical competition. In David Lightfoot (ed.), Syntactic Effects of Morphological Change, 276–299. Oxford: Oxford University Press.

Pintzuk, Susan and Ann Taylor. 2006. The loss of OV order in the history of English. In Ans van Kemenade and Bettelou Los (eds.), Blackwell Handbook of the History of English, 249–278. Oxford: Blackwell Publishing.

Pyles, Thomas. 1971. The Origins and Development of the English Language. 2nd ed. New York: Harcourt Brace Jovanovich.

Roberts, Ian. 1997. Directionality and word order change in the history of English. In Ans van Kemenade and Nigel Vincent (eds.), Parameters of Morphosyntactic Change, 397–426. Cambridge: Cambridge University Press.

Stockwell, Robert and Donca Minkova. 1991. Subordination and word order change in the history of English. In Dieter Kastovsky (ed.), Historical English Syntax, 367–408. Berlin: Mouton de Gruyter.

Traugott, Elisabeth C. 1972. A History of English Syntax. A Transformational Approach to the History of English Sentence Structure. New York: Holt, Rinehart and Winston.